The plate has the following dimensions : 30 x 20 x 3 [mm]
The two holes have once 9.6mm and once 11.9mm and 9.6mm as diameter.
I designed in Fusion 360 and selected high resolution for the .stl export/save
The print took 18min ( mid quality) and 10min ( low quality).
Results: Both two pieces looked very similar and I can’t really distinguish them, thus i write only the dimensions of one sample:
Plate : 30.04 x 20.05 x3.03 [mm]
Holes: 11.6mm and 9.2mm
My question to you is: Why is the accuracy of the plate such good and in comparision such bad on these holes?
Did I expect too much? My goal was to pass a bolt trough the holes with the diameter 11.7 and 9.4.
Is the margin of 0.2mm to low? What would you recommend ?
As a new meber of the forum I can’t upload an attachement I will look how i can share with you.
When you write excellent, what does it mean exactly ?
Edit:
I don’t know if external links like this are allowed, and honestly, i don’t know this site but it was just the first link for free sharing platform showed up by google.
Excellent is subjective sure: to me that means my machine has backlash compensation enabled and I can command it to move to 0.02mm repeatability, as measured with a dial indicator. On a 20mm calibration cube the dimensions as measured at each edge and middle are as follows:
On a single walled vase-mode printed cube, with a commanded wall thickness in the slicer of 0.50mm, I see +/-0.05mm deviation from that.
For parts that must fit together I design using a 0.1mm/0.15mm/0.2mm clearance for interference/transitional/clearance fits, and the printed dimensions I get work with those clearances.
I have the file, that works. Give me a minute with that.
Your part was as you described, but I wasn’t envisioning it until I saw the model. Visuals help.
I would expect the hole with the diameter of 9.6mm to pass anything less than 9.4mm through it. And the shoulder of 11.9, 11.7mm to fit in that. Typically 0.2mm clearance fits work for me.
Since you’re getting .3-.4mm smaller than that I’m going to assume you have overextrusion going on that is forcing extra filament into the inner diameter of the holes, reducing clearence.
For a test piece, it’s easy enough to modify slicer settings and try again. For a part that I can’t afford to waste filament on I’ll ream it to final dimension on the drill press using a carbide burr or drill bit.
Parts where the inside must be very smooth I will deliberately undersize and ream to final dimension, also. I did that with some bearings I printed to precisely mate with a slip fit onto a steel shaft.
So i tryed to reduce the extrusion ( 90% , 80%, 60%) but none of them helped really execpt the 60%. There I measured 9.4/9.45 and 11.9mm. But the extrusion was so low that the quality dropped extremely
With 80% it already startet to have a little gap between the walls and the rest of the layers.
So possibly with 90% I can continue.
Are there other settings which could help to resolve the problems?
Your extrusion multiplier should hover around 90-100% unless you have a specific need and are aware of the consequences of deviating far from that. As you found, quality is extremely sensitive to changes in that.
And yea, possibly too many settings, and they all can interact with each other a bit. I’ll list settings in Simplify3D, many exist in Cura with similar or different names. Many of these do not exist in Luban, but are important anyways, and are a reason people should consider not using Luban to slice. The ones that are bold are the important ones to focus on, the rest should be checked if you are not happy with quality. This list is long because it’s intended to be more of a reference rather than a response to your specific problem.
Main ones:
Machine, in roughly descending importance:
Extruder Steps/mm (e-steps). I’d recommend doing this repeatedly until you can get 2 or 3 100mm marked filament lengths to pretty accurately. If you cannot get known filament feeding lengths, everything else will fail to be meaningful adjustments and will just be blindly turning knobs. Cannot overstate the importance of having correct e-steps.
Filament Diameter (measure with a micrometer. 1.72mm vs 1.75mm will have 4% over/under extrusion.
Extruder Temperature (in the sense that too hot can cause filament to sag and run more than it should normally)
Linear advance. Depending on your print speed and the specific feature being printed this can have drastic effects, or none at all.
Acceleration, Junction Deviation (More overall surface finish quality than dimensional, but if it’s way out of wack it can affect dimensional accuracy if printed at high speed).
Slicer, in roughly descending importance:
Filament diameter (same as above - needs to be set accurately, very important)
Extrusion/Flow Multiplier - Used when setting e-steps and filament diameter can’t adequately control the output. Can be overriden temporarily mid-print with M221.
Number of solid top / bottom layers. With an insufficient number of top solid layers the top will underextrude. Print a solid test object at 100% infill, like a 10x10x10mm cube. Ideally, it should look the same as one printed with 10% infill. If there are too few solid top layers, the first few layers printed on top of the sparse infill will sag and be underextruded due to spanning free air and not being compressed. There needs to be enough layers that allows a flat base to be built up to be dimensionally accurate - at least 3 at minimum, but with thin layer heights possibly 6 or more.
Infill overlap (with the outer perimeter) - This also affect solid layers, which are a special type of infill. If there is too much overlap extra filament will be pushed up causing overextrusion issues. But if it’s too low there will be gaps at the boundary between the solid infill and perimeter.
Infill density - related to above # of solid layers in the sense that if you cannot get adequate solid layers at spare infill then you need to increase the infill density from maybe 10% to 25%, part and feature dependent. This also has strength implications, as 100% infill will be the strongest (completely solid part).
Thin wall behavior (single width extrusion variable gap fill). Depending on your part geometry you can end up with small gaps here and there and if a slicer can vary the extrusion width you can ensure those gaps get filled without causing overextrusion. This is less about dimensional accuracy, and more improving quality after tweaking some of the other settings.
Additionally for part strength, more perimeters is stronger, up to a point. For “structural” prints I’ve used 100% infill with 5 perimeters.
I find it helpful to think about all of these settings in terms of which ones are “physical”, in the sense that they need to match real world measured values, and which ones are “tweaks”, in the sense that they can be subjectively changed. E-steps and filament diameter are real physical quantities that need to match reality, and should not be “tweaked” to fix a problem like over or under extrusion. The tweaking should be done in extrusion multiplier, like you mentioned you were doing.
Since that previous post was long and general, just wanted to specifically mention what you said here.
With a gap between the perimeter and fill, that can be adjusted with the infill overlap settings, without affecting (too much) the rest of the extrusion quality.
In other words, if the first layers print great, the walls print great, and top layer has great surface quality, and the only problem is the solid layers are not well bonded to the perimeter, then infill overlap is the setting to adjust.
Cura has good documentation on infill settings here: Support Community
I started with measuring the esteps. I put a mark 110mm away from the extruder and sent the command “G1 E100 F100” which i found online ( 100mm in one minute). The I measured the mark with a result of 22mmm which would mean 12mm to short.
Is this a normal deviation before calibrating or am I doing something wrong? Is this normal for a Snapmaker 2 A350?
If this is normal, how could I be happy with my results of previous prints?
I haven’t modified anything for the e-steps yet, will wait your answer to the plausibility of the result.
Then:
I measured the filament, not really over a huge distance, but received diameters betweene 1.77 and 1.78mm. I my understanding i should decrease my extrusion by appr. 3%.
The next thing i have done is the 10m cube with 100% infill, ( not shure if the result is useful due to the e-step measurement ) but:
z = 10.08mm
y = 9.95mm
x = 9.96mm
I had also a little little “elephant foot” but if I switch to cura probably i will find a parameter to decrease the flow for the first layer. So I could even lover the z axis by 0.05 to be fine.
Yes, unfortunately. Most people receive their machines with significant error in the e steps. Fixing that and the filament diameter is the foundation to build other calibrations on top of.
Arg, how frustrating. I didn’t realize Luban will not allow you to change filament diameter. Cura and S3D both allow you to tweak the filament diameter, and it calculates the extrusion change for you. Since Luban uses Cura to slice, under the hood, it’s a pretty straightforward change to start using Cura, and it has many more helpful tools for dialing in prints. In the meantime though, yes, your flow percentage will have to be compensated by -3% +/- any additional amount you find is necessary.
I’d recommend making a note or something because next time you change filaments you’ll need to change the 3% part of that number. For example, if you end up having to make an additional tweak to the flow to dial in your wall thickness, let’s say you end up at -7% total. Then you get a new spool of filament that’s 1.75mm. You’ll need to back out the -3% from the diameter change, but leave the other -4% because that’s tied to PLA material in general. That’s why having separate numbers is better.
Overall, that’s a good looking print. Glad you were able to get such a good print with only the basic checks. If you run into other issues, there’s more knobs to turn, so to speak.
Usually elephants foot is, as you said, first layer height/flow related. You might be able to start your print a small bit higher in Z, by dialing in a +0.05mm z offset in the controller. There’s other ways too. And if you have a file or rasp, sometimes having a brim or bit of elephant foot helps with adhesion and you can remove it later with a file.
This is good stuff here. So much assembled knowledge presented in a meaningful way
Couple other questions - In cura, do you go into the extruder settings and change the value, or just adjust the flow rate?
or just
(plus the various sub settings)
Also, I have to look into this estep thing, i did the extruder calibration (i probably should do it again with more precision) but didnt have a way that printed a line to measure after the fact, just measured how much was consumed through the hotend and measuring what was left from a 100 mm piece.
Furthermore, you say “leave the other -4% because that’s tied to PLA material in general” - Is this a thing to know or is this circumstantial? Does telling Cura you are using PLA make this adjustment automatically or do I have to adjust the flow rate by 4% by default?
Personally, I would check the material diameter every time I change filament spools and update. Everything else should stay the same, if the material stays the same.
No, you need to choose a value based on your print quality
This is a circumstantial value tied to a specific machine and filament. Here’s a couple resources that talk more about it:
On the Teaching Tech Flow Calibration Tools page he mentions when changing the flow should be required: Base calibration, as well as any time there has been a change to the extruder/hot end.
The Simplify3d support pages have lots of great information. The Under Extrusion Troubleshooting page mentions the following: Each extruder on your printer can have a unique extrusion multiplier. It is typical for PLA to print with an extrusion multiplier near 0.9, while ABS tends to have extrusion multipliers closer to 1.0.
With regard to the ‘PLA is typically 0.9’, I haven’t experienced that, I print at 1.0 with PLA. But I know many people that do set near 0.9, so it needs to be something you determine for yourself using a test print like the above linked Teaching Tech website.
Okay so I should go into the extruder settings and adjust the filament diameter based on the measurement (in several places) is what you are saying. This spool on my rig now is 1.74, and im about to start a new job so im gona try it!
This comes from people printing pla at higher temps. Alot of people print pla at ~205 at that temp it flows easier so your multiplier needs to be lower. The advantage is better layer adhesion. If you print closer to 195 then you will find a multiplier of 1 will work better
I will look how i can share with you.
